Fellow Chrom-ers
We are running an isocratic RP-C18 analysis of Leukotriene metabolites using Millenium ver4.2.
The number if tests within a sample set is about 35 & the system P is typically 2800psi. During the execution of the sample set, the system exceeds the max system pressure 4000 psi and automatically shutdown. I can't explain the primary symptom 'High Pressure' leading to the shutdown. This has occured on 3 individual sample sets however I am able to continue the sample set analysis by slow manual increments of the flow after loosening the fitting of the column inlet or guard inlet. Sample set is restarted (at vial incompleted) and goes to completion.

the system used is:
Water System: 717 + 600 + 996pda
Column: Waters Symm C18
MP: 70 Meoh + 33 H2O + 0.04 NH4OH pH3
NOTE: Meoh & H20 individually filtered & then mixed thoroughly
Flow: 1 ml/min
System psi: 2800-3000psi
Chromatography: peaks symmetry OK,

My approach so far has been:
1) Checked for blockages- loosen at column inlet, guard inlet
2) Checked MP (outgassing) Performed compressibility check on 717 (OK) and purged pump
3) Transducer- reset system psi at 60 f=0 ml/min

Following check is to plate test the column however the column is relatively new ie total 450 injections.

Any ideas on the sharp increase in pressure leading to shutdown?

addendum: We have included monitoring of the systme pressure for each rund executed in the sample set. We have observed that the pressure is stable @ 2950 however there is a sharp -ve spike (-700psi) @ 17min prior to Meoh wash @19min. So far, the psi dip-spike appears on every 3rd test run in sequence.

There have been quite a number of chains on plugging, etc., of columns, so what is the matrix of your samples (what do you inject)? Is it the "old" macro-molecule problem?

Another general appeal: Please give more info on matters in direct causative relationship with the problem.

Also, is this correct that you use a diode array detector? We didnīt see several leukotrienes (not to even talk of metabolites, usually did see LTC4, though) in plasma/serum even with a highly sensitive normal UV detector (used ELISA for detection, better: rough estimation). If we concentrated samples more we got a continuum of "dirt", even after three columns in series (the first column was a restricted access 4.6x250mm Pinkerton).

Set up a sample ser that injects mobile phase. Does the same problem occur? This simple experiment eliminates (or confirms) that the sample is the cause of this issue. If the problem still occurs, then start looking at the method it self. You talk about a MeOH wash, precipitation at this wash step is a possibility. It is often better to filter pre-mixed mobile phases after mixing (to eliminate anything that may drop out of solution upon mixing).

What is 0.04 NH4OH pH 3?

Good point anon. What acid did you use to adjust the pH of your ammonium hydroxide. Hopefully not HCl as this is rather corrosive and could be the source of your problems.

Maybe he/she means NH4Ac rather than NH4OH? At that pH is would make more sense to use plain HOAc, but it's a thought...

By nic on Monday, September 9, 2002 - 06:25 pm:
Thanks to 'all' for the trouble-shooting leads. In reply to H W Mueller: Sample matrix is a biological. I've used incorrect terms: apologies. We analyse LPO metabolites ie LTB4 & its trans isomers 6t-LTB4 , 6t 12 epi-LTB4 (non-enzymic) plus 5-HETE from porcine neutrophils stimulated by the addition of arachidonic acid/calcium ionophore (method adapted from McColl et al, JChrom 1986, 378, 444-449). Samples are extracted, evaporated to dryness & reconstituted in mobile phase. LPO metabolites confirmed with commercial stds: LTBs @270nm & HETE @235nm. Others have reported separation of LTC4, LTB4 @280nm.

To Anonymous & juddc: I resubmit the mobile phase components:
LTB4 mobile phase in parts: 76MeOH + 34H2O + 0.08 HOAc pH3
HETE mobile phase in parts: 80MeOH + 30H2O + 0.08HOAc + 0.08NH4OAc pH6.2

Generally, we use acetic acid or TFA to adjust pH. Point taken (re: Anon) on the importance of filtering pre-mixed mobile phase. Acetic acid is sol in mobile phase & MeOH. I've observed the percipitation effect when using aqueous MeOH phosphate buffered mobile phase.

Hope to have answered the 'message' gaps. I will try the blank sample set.

On the sensitivity: Ok, you seem to be lucky not to have to work with human plasma/serum, your system is defined much better.
On the pressure problem: If you do not regularily (at least every evening)clean your column with methanol or another organic, do so. But donīt let pure organic sit on the column over night, it seems to "cement" remaining proteins in place. Maybe a MeOH/H2O mix similar to your mobile phase, without buffer..., is approriate. If it doesnīt work wash the column with urea (or other chaotropics), allways backwards if the column allows! Still no go? Try an aqu. solution of lithium dodecylsulfate and dithiothreitol (a lengthy process). It may be advantageous to change the entrance frit before trying the latter (careful!).
If you are using a guard column or pre-filter you may need to change or clean only those. A daily org. wash (guard or pre-frit separatly) will be necessary in any case.
Anyway it would be nice to hear how you solved the problem.

thanks H W M
I'll include the column wash house-keeping, daily. The term 'chaotropic' is new to me. pls define.
I'll keep u posted.

Also due to post a NP-HPLC + ELSD problem for your feed-back.

yia

A crude, but workable definition of chaotropic substances: Materials which interfere with the structure of water (for instance they lower the surface tension) so as to make it to be (or appear to be) less polar. Therefore, chaotropic substances have the tendency to increase the solubility of hydrophobic compounds in H2O. The opposite is lyotropic. There exist compilations in books on bio-methods, proteins, etc., about chaotropic or lyotropic salt series, for instance, NH4SO4 is on the lyotropic (low chaotropic)end of such series and is commonly used to precipitate proteins. NaSCN is on the chaotropic side and can help to solvate proteins (a bit oversimplified).
Now it would be very helpful if someone could answer the question on how the chaotropes, urea, guanidine HCl, etc., fit into these series and why organic solvents with similar properties are not classified as chaotropic/lyotropic. Also, I am not sure as to why detergents (another group of compounds that figure in H2O-protein interaction) should not be used simultaneously with chaotropes.

Update on the 'system-pressure-shutdown' phenomenon.

I took Anonymous's lead & set-up a sample set that injects the mobile phase. Mock-sample set_1 did not run to completion: system shut-down. The psi profiles were stable @3000 +/- 100 psi during the course however at vial 17, psi exceeded 4000 psi @11 min within 1 min interval.

I observed that the PDA status indicator light was flashing and recalled that before execution of the Mock sample set, the light was in steady state. This suggests that there was air in the flow cell or the lamp has failed. I powered the pda off, flushed the cell with fresh mobile phase, then powered on pda, waited one hour and ran pda diagnostics. All checked OK.
The pda manual recommends a 30-60cm of 9 thou ID tube on the pda waste outlet. I changed the tube outlet as recommended and ran Mock sample set 2 with fresh mobile phase. The run went to completion: the pressure profiles were stable. Following that, the system has been OK. My conclusion to date is that the air lock in the pda cell has lead to the system shutdown. Your thought pls.

The pressure profiles on all runs show an initial psi drop of 200 psi at the start of each injection. how can this be explained? 717 pressure pulses are more sensitive with RI than UV or PDA

Air 'locks' in detector cells won't cause pressures to go sky high. If your system is shutting down, it must be for some other reason. However, by plotting the pressure during the run, and discovering the 4K spike mid-run, you confirmed a flow obstruction--somewhere.

You checked the manual for the waste line specifications, changed the detector waste line to meet the specs, and now everything is fine. So now where do you think the obstruction was?

A good rule to follow is when you get a waste line system that works with a detector, DO NOT CHANGE IT!! I've seen lots of incidents over the years with chemists taking lines from detector A, putting them on B, and having trouble, including blowing up cells. I discovered this problem early in my LC career, and can't recall an incident since.

There are plenty of other things to go wrong in an LC, without having to worry about this easily avoided issue as well.

Don't know if you are still having this problem. Does the problem always occur at vial 17? What happens if you start a run at vial 17? I am not familiar with the Waters system. If the injection is made through a multi-port valve which gets rotated on injection, does the initial pressure drop occur when the sample is loaded onto the loop or after the loop is rotated to inject the sample onto the column? Could you be sporadically losing the home position of the injection valve? If it does not rotate to the correct position, a blockage will result and the pressure will go up.

Hi nic,
do you use a degasser with your hplc-system?
Is the silk-option enabled in your instrument method?
With the silk-option enabled, even small bubbles reaching the pump could lead to high-pressure and if the pressure exceeds the limits, shutdown occurs...

I'm working with the 717+/616/996 (plus inline degasser) system since 1994 and couldn't remember having the same problem as you described (and I saw some real strange things happen...).
When the 717+ injects, you always see pressure up/down coming from the (pressureless) sample loop or the switching of the valves when injection occurs. But with an high pressure limit 1000 psi over work pressure this should be no problem.

The flashing light on the 996 indicates normaly ear bubbles in the detector cell (coming from a run dry of mobile phase or changing the column). Only once I managed to reinstall in a total wrong way the cell so no light was coming through.
That could indicate that bubbles were coming through the solvent into the system and finaly led to high pressure.

If you use an inline degasser, check out the screws of the solvent tubes, maybe there loose.